TAPAS: a Toolbox for Adversarial Privacy Auditing of Synthetic Data
Abstract
Personal data collected at scale promises to improve decision-making and accelerate innovation. However, sharing and using such data raises serious privacy concerns. A promising solution is to produce synthetic data, artificial records to share instead of real data. Since synthetic records are not linked to real persons, this intuitively prevents classical re-identification attacks. However, this is insufficient to protect privacy. We here present TAPAS, a toolbox of attacks to evaluate synthetic data privacy under a wide range of scenarios. These attacks include generalizations of prior works and novel attacks. We also introduce a general framework for reasoning about privacy threats to synthetic data and showcase TAPAS on several examples.
Keywords
Cite
@article{arxiv.2211.06550,
title = {TAPAS: a Toolbox for Adversarial Privacy Auditing of Synthetic Data},
author = {Florimond Houssiau and James Jordon and Samuel N. Cohen and Owen Daniel and Andrew Elliott and James Geddes and Callum Mole and Camila Rangel-Smith and Lukasz Szpruch},
journal= {arXiv preprint arXiv:2211.06550},
year = {2022}
}
Comments
Published at the SyntheticData4ML Neurips workshop